1# SPDX-License-Identifier: GPL-2.0-only 2config CC_VERSION_TEXT 3 string 4 default "$(CC_VERSION_TEXT)" 5 help 6 This is used in unclear ways: 7 8 - Re-run Kconfig when the compiler is updated 9 The 'default' property references the environment variable, 10 CC_VERSION_TEXT so it is recorded in include/config/auto.conf.cmd. 11 When the compiler is updated, Kconfig will be invoked. 12 13 - Ensure full rebuild when the compiler is updated 14 include/linux/compiler-version.h contains this option in the comment 15 line so fixdep adds include/config/CC_VERSION_TEXT into the 16 auto-generated dependency. When the compiler is updated, syncconfig 17 will touch it and then every file will be rebuilt. 18 19config CC_IS_GCC 20 def_bool $(success,test "$(cc-name)" = GCC) 21 22config GCC_VERSION 23 int 24 default $(cc-version) if CC_IS_GCC 25 default 0 26 27config CC_IS_CLANG 28 def_bool $(success,test "$(cc-name)" = Clang) 29 30config CLANG_VERSION 31 int 32 default $(cc-version) if CC_IS_CLANG 33 default 0 34 35config AS_IS_GNU 36 def_bool $(success,test "$(as-name)" = GNU) 37 38config AS_IS_LLVM 39 def_bool $(success,test "$(as-name)" = LLVM) 40 41config AS_VERSION 42 int 43 # Use clang version if this is the integrated assembler 44 default CLANG_VERSION if AS_IS_LLVM 45 default $(as-version) 46 47config LD_IS_BFD 48 def_bool $(success,test "$(ld-name)" = BFD) 49 50config LD_VERSION 51 int 52 default $(ld-version) if LD_IS_BFD 53 default 0 54 55config LD_IS_LLD 56 def_bool $(success,test "$(ld-name)" = LLD) 57 58config LLD_VERSION 59 int 60 default $(ld-version) if LD_IS_LLD 61 default 0 62 63config CC_CAN_LINK 64 bool 65 default $(success,$(srctree)/scripts/cc-can-link.sh $(CC) $(CLANG_FLAGS) $(USERCFLAGS) $(USERLDFLAGS) $(m64-flag)) if 64BIT 66 default $(success,$(srctree)/scripts/cc-can-link.sh $(CC) $(CLANG_FLAGS) $(USERCFLAGS) $(USERLDFLAGS) $(m32-flag)) 67 68config CC_CAN_LINK_STATIC 69 bool 70 default $(success,$(srctree)/scripts/cc-can-link.sh $(CC) $(CLANG_FLAGS) $(USERCFLAGS) $(USERLDFLAGS) $(m64-flag) -static) if 64BIT 71 default $(success,$(srctree)/scripts/cc-can-link.sh $(CC) $(CLANG_FLAGS) $(USERCFLAGS) $(USERLDFLAGS) $(m32-flag) -static) 72 73config CC_HAS_ASM_GOTO 74 def_bool $(success,$(srctree)/scripts/gcc-goto.sh $(CC)) 75 76config CC_HAS_ASM_GOTO_OUTPUT 77 depends on CC_HAS_ASM_GOTO 78 def_bool $(success,echo 'int foo(int x) { asm goto ("": "=r"(x) ::: bar); return x; bar: return 0; }' | $(CC) -x c - -c -o /dev/null) 79 80config CC_HAS_ASM_GOTO_TIED_OUTPUT 81 depends on CC_HAS_ASM_GOTO_OUTPUT 82 # Detect buggy gcc and clang, fixed in gcc-11 clang-14. 83 def_bool $(success,echo 'int foo(int *x) { asm goto (".long (%l[bar]) - .\n": "+m"(*x) ::: bar); return *x; bar: return 0; }' | $CC -x c - -c -o /dev/null) 84 85config TOOLS_SUPPORT_RELR 86 def_bool $(success,env "CC=$(CC)" "LD=$(LD)" "NM=$(NM)" "OBJCOPY=$(OBJCOPY)" $(srctree)/scripts/tools-support-relr.sh) 87 88config CC_HAS_ASM_INLINE 89 def_bool $(success,echo 'void foo(void) { asm inline (""); }' | $(CC) -x c - -c -o /dev/null) 90 91config CC_HAS_NO_PROFILE_FN_ATTR 92 def_bool $(success,echo '__attribute__((no_profile_instrument_function)) int x();' | $(CC) -x c - -c -o /dev/null -Werror) 93 94config PAHOLE_VERSION 95 int 96 default $(shell,$(srctree)/scripts/pahole-version.sh $(PAHOLE)) 97 98config CONSTRUCTORS 99 bool 100 101config IRQ_WORK 102 bool 103 104config BUILDTIME_TABLE_SORT 105 bool 106 107config THREAD_INFO_IN_TASK 108 bool 109 help 110 Select this to move thread_info off the stack into task_struct. To 111 make this work, an arch will need to remove all thread_info fields 112 except flags and fix any runtime bugs. 113 114 One subtle change that will be needed is to use try_get_task_stack() 115 and put_task_stack() in save_thread_stack_tsk() and get_wchan(). 116 117menu "General setup" 118 119config BROKEN 120 bool 121 122config BROKEN_ON_SMP 123 bool 124 depends on BROKEN || !SMP 125 default y 126 127config INIT_ENV_ARG_LIMIT 128 int 129 default 32 if !UML 130 default 128 if UML 131 help 132 Maximum of each of the number of arguments and environment 133 variables passed to init from the kernel command line. 134 135config COMPILE_TEST 136 bool "Compile also drivers which will not load" 137 depends on HAS_IOMEM 138 help 139 Some drivers can be compiled on a different platform than they are 140 intended to be run on. Despite they cannot be loaded there (or even 141 when they load they cannot be used due to missing HW support), 142 developers still, opposing to distributors, might want to build such 143 drivers to compile-test them. 144 145 If you are a developer and want to build everything available, say Y 146 here. If you are a user/distributor, say N here to exclude useless 147 drivers to be distributed. 148 149config WERROR 150 bool "Compile the kernel with warnings as errors" 151 default COMPILE_TEST 152 help 153 A kernel build should not cause any compiler warnings, and this 154 enables the '-Werror' flag to enforce that rule by default. 155 156 However, if you have a new (or very old) compiler with odd and 157 unusual warnings, or you have some architecture with problems, 158 you may need to disable this config option in order to 159 successfully build the kernel. 160 161 If in doubt, say Y. 162 163config UAPI_HEADER_TEST 164 bool "Compile test UAPI headers" 165 depends on HEADERS_INSTALL && CC_CAN_LINK 166 help 167 Compile test headers exported to user-space to ensure they are 168 self-contained, i.e. compilable as standalone units. 169 170 If you are a developer or tester and want to ensure the exported 171 headers are self-contained, say Y here. Otherwise, choose N. 172 173config LOCALVERSION 174 string "Local version - append to kernel release" 175 help 176 Append an extra string to the end of your kernel version. 177 This will show up when you type uname, for example. 178 The string you set here will be appended after the contents of 179 any files with a filename matching localversion* in your 180 object and source tree, in that order. Your total string can 181 be a maximum of 64 characters. 182 183config LOCALVERSION_AUTO 184 bool "Automatically append version information to the version string" 185 default y 186 depends on !COMPILE_TEST 187 help 188 This will try to automatically determine if the current tree is a 189 release tree by looking for git tags that belong to the current 190 top of tree revision. 191 192 A string of the format -gxxxxxxxx will be added to the localversion 193 if a git-based tree is found. The string generated by this will be 194 appended after any matching localversion* files, and after the value 195 set in CONFIG_LOCALVERSION. 196 197 (The actual string used here is the first eight characters produced 198 by running the command: 199 200 $ git rev-parse --verify HEAD 201 202 which is done within the script "scripts/setlocalversion".) 203 204config BUILD_SALT 205 string "Build ID Salt" 206 default "" 207 help 208 The build ID is used to link binaries and their debug info. Setting 209 this option will use the value in the calculation of the build id. 210 This is mostly useful for distributions which want to ensure the 211 build is unique between builds. It's safe to leave the default. 212 213config HAVE_KERNEL_GZIP 214 bool 215 216config HAVE_KERNEL_BZIP2 217 bool 218 219config HAVE_KERNEL_LZMA 220 bool 221 222config HAVE_KERNEL_XZ 223 bool 224 225config HAVE_KERNEL_LZO 226 bool 227 228config HAVE_KERNEL_LZ4 229 bool 230 231config HAVE_KERNEL_ZSTD 232 bool 233 234config HAVE_KERNEL_UNCOMPRESSED 235 bool 236 237choice 238 prompt "Kernel compression mode" 239 default KERNEL_GZIP 240 depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA || HAVE_KERNEL_XZ || HAVE_KERNEL_LZO || HAVE_KERNEL_LZ4 || HAVE_KERNEL_ZSTD || HAVE_KERNEL_UNCOMPRESSED 241 help 242 The linux kernel is a kind of self-extracting executable. 243 Several compression algorithms are available, which differ 244 in efficiency, compression and decompression speed. 245 Compression speed is only relevant when building a kernel. 246 Decompression speed is relevant at each boot. 247 248 If you have any problems with bzip2 or lzma compressed 249 kernels, mail me (Alain Knaff) <alain@knaff.lu>. (An older 250 version of this functionality (bzip2 only), for 2.4, was 251 supplied by Christian Ludwig) 252 253 High compression options are mostly useful for users, who 254 are low on disk space (embedded systems), but for whom ram 255 size matters less. 256 257 If in doubt, select 'gzip' 258 259config KERNEL_GZIP 260 bool "Gzip" 261 depends on HAVE_KERNEL_GZIP 262 help 263 The old and tried gzip compression. It provides a good balance 264 between compression ratio and decompression speed. 265 266config KERNEL_BZIP2 267 bool "Bzip2" 268 depends on HAVE_KERNEL_BZIP2 269 help 270 Its compression ratio and speed is intermediate. 271 Decompression speed is slowest among the choices. The kernel 272 size is about 10% smaller with bzip2, in comparison to gzip. 273 Bzip2 uses a large amount of memory. For modern kernels you 274 will need at least 8MB RAM or more for booting. 275 276config KERNEL_LZMA 277 bool "LZMA" 278 depends on HAVE_KERNEL_LZMA 279 help 280 This compression algorithm's ratio is best. Decompression speed 281 is between gzip and bzip2. Compression is slowest. 282 The kernel size is about 33% smaller with LZMA in comparison to gzip. 283 284config KERNEL_XZ 285 bool "XZ" 286 depends on HAVE_KERNEL_XZ 287 help 288 XZ uses the LZMA2 algorithm and instruction set specific 289 BCJ filters which can improve compression ratio of executable 290 code. The size of the kernel is about 30% smaller with XZ in 291 comparison to gzip. On architectures for which there is a BCJ 292 filter (i386, x86_64, ARM, IA-64, PowerPC, and SPARC), XZ 293 will create a few percent smaller kernel than plain LZMA. 294 295 The speed is about the same as with LZMA: The decompression 296 speed of XZ is better than that of bzip2 but worse than gzip 297 and LZO. Compression is slow. 298 299config KERNEL_LZO 300 bool "LZO" 301 depends on HAVE_KERNEL_LZO 302 help 303 Its compression ratio is the poorest among the choices. The kernel 304 size is about 10% bigger than gzip; however its speed 305 (both compression and decompression) is the fastest. 306 307config KERNEL_LZ4 308 bool "LZ4" 309 depends on HAVE_KERNEL_LZ4 310 help 311 LZ4 is an LZ77-type compressor with a fixed, byte-oriented encoding. 312 A preliminary version of LZ4 de/compression tool is available at 313 <https://code.google.com/p/lz4/>. 314 315 Its compression ratio is worse than LZO. The size of the kernel 316 is about 8% bigger than LZO. But the decompression speed is 317 faster than LZO. 318 319config KERNEL_ZSTD 320 bool "ZSTD" 321 depends on HAVE_KERNEL_ZSTD 322 help 323 ZSTD is a compression algorithm targeting intermediate compression 324 with fast decompression speed. It will compress better than GZIP and 325 decompress around the same speed as LZO, but slower than LZ4. You 326 will need at least 192 KB RAM or more for booting. The zstd command 327 line tool is required for compression. 328 329config KERNEL_UNCOMPRESSED 330 bool "None" 331 depends on HAVE_KERNEL_UNCOMPRESSED 332 help 333 Produce uncompressed kernel image. This option is usually not what 334 you want. It is useful for debugging the kernel in slow simulation 335 environments, where decompressing and moving the kernel is awfully 336 slow. This option allows early boot code to skip the decompressor 337 and jump right at uncompressed kernel image. 338 339endchoice 340 341config DEFAULT_INIT 342 string "Default init path" 343 default "" 344 help 345 This option determines the default init for the system if no init= 346 option is passed on the kernel command line. If the requested path is 347 not present, we will still then move on to attempting further 348 locations (e.g. /sbin/init, etc). If this is empty, we will just use 349 the fallback list when init= is not passed. 350 351config DEFAULT_HOSTNAME 352 string "Default hostname" 353 default "(none)" 354 help 355 This option determines the default system hostname before userspace 356 calls sethostname(2). The kernel traditionally uses "(none)" here, 357 but you may wish to use a different default here to make a minimal 358 system more usable with less configuration. 359 360config SYSVIPC 361 bool "System V IPC" 362 help 363 Inter Process Communication is a suite of library functions and 364 system calls which let processes (running programs) synchronize and 365 exchange information. It is generally considered to be a good thing, 366 and some programs won't run unless you say Y here. In particular, if 367 you want to run the DOS emulator dosemu under Linux (read the 368 DOSEMU-HOWTO, available from <http://www.tldp.org/docs.html#howto>), 369 you'll need to say Y here. 370 371 You can find documentation about IPC with "info ipc" and also in 372 section 6.4 of the Linux Programmer's Guide, available from 373 <http://www.tldp.org/guides.html>. 374 375config SYSVIPC_SYSCTL 376 bool 377 depends on SYSVIPC 378 depends on SYSCTL 379 default y 380 381config POSIX_MQUEUE 382 bool "POSIX Message Queues" 383 depends on NET 384 help 385 POSIX variant of message queues is a part of IPC. In POSIX message 386 queues every message has a priority which decides about succession 387 of receiving it by a process. If you want to compile and run 388 programs written e.g. for Solaris with use of its POSIX message 389 queues (functions mq_*) say Y here. 390 391 POSIX message queues are visible as a filesystem called 'mqueue' 392 and can be mounted somewhere if you want to do filesystem 393 operations on message queues. 394 395 If unsure, say Y. 396 397config POSIX_MQUEUE_SYSCTL 398 bool 399 depends on POSIX_MQUEUE 400 depends on SYSCTL 401 default y 402 403config WATCH_QUEUE 404 bool "General notification queue" 405 default n 406 help 407 408 This is a general notification queue for the kernel to pass events to 409 userspace by splicing them into pipes. It can be used in conjunction 410 with watches for key/keyring change notifications and device 411 notifications. 412 413 See Documentation/watch_queue.rst 414 415config CROSS_MEMORY_ATTACH 416 bool "Enable process_vm_readv/writev syscalls" 417 depends on MMU 418 default y 419 help 420 Enabling this option adds the system calls process_vm_readv and 421 process_vm_writev which allow a process with the correct privileges 422 to directly read from or write to another process' address space. 423 See the man page for more details. 424 425config USELIB 426 bool "uselib syscall (for libc5 and earlier)" 427 default ALPHA || M68K || SPARC 428 help 429 This option enables the uselib syscall, a system call used in the 430 dynamic linker from libc5 and earlier. glibc does not use this 431 system call. If you intend to run programs built on libc5 or 432 earlier, you may need to enable this syscall. Current systems 433 running glibc can safely disable this. 434 435config AUDIT 436 bool "Auditing support" 437 depends on NET 438 help 439 Enable auditing infrastructure that can be used with another 440 kernel subsystem, such as SELinux (which requires this for 441 logging of avc messages output). System call auditing is included 442 on architectures which support it. 443 444config HAVE_ARCH_AUDITSYSCALL 445 bool 446 447config AUDITSYSCALL 448 def_bool y 449 depends on AUDIT && HAVE_ARCH_AUDITSYSCALL 450 select FSNOTIFY 451 452source "kernel/irq/Kconfig" 453source "kernel/time/Kconfig" 454source "kernel/bpf/Kconfig" 455source "kernel/Kconfig.preempt" 456 457menu "CPU/Task time and stats accounting" 458 459config VIRT_CPU_ACCOUNTING 460 bool 461 462choice 463 prompt "Cputime accounting" 464 default TICK_CPU_ACCOUNTING if !PPC64 465 default VIRT_CPU_ACCOUNTING_NATIVE if PPC64 466 467# Kind of a stub config for the pure tick based cputime accounting 468config TICK_CPU_ACCOUNTING 469 bool "Simple tick based cputime accounting" 470 depends on !S390 && !NO_HZ_FULL 471 help 472 This is the basic tick based cputime accounting that maintains 473 statistics about user, system and idle time spent on per jiffies 474 granularity. 475 476 If unsure, say Y. 477 478config VIRT_CPU_ACCOUNTING_NATIVE 479 bool "Deterministic task and CPU time accounting" 480 depends on HAVE_VIRT_CPU_ACCOUNTING && !NO_HZ_FULL 481 select VIRT_CPU_ACCOUNTING 482 help 483 Select this option to enable more accurate task and CPU time 484 accounting. This is done by reading a CPU counter on each 485 kernel entry and exit and on transitions within the kernel 486 between system, softirq and hardirq state, so there is a 487 small performance impact. In the case of s390 or IBM POWER > 5, 488 this also enables accounting of stolen time on logically-partitioned 489 systems. 490 491config VIRT_CPU_ACCOUNTING_GEN 492 bool "Full dynticks CPU time accounting" 493 depends on HAVE_CONTEXT_TRACKING 494 depends on HAVE_VIRT_CPU_ACCOUNTING_GEN 495 depends on GENERIC_CLOCKEVENTS 496 select VIRT_CPU_ACCOUNTING 497 select CONTEXT_TRACKING 498 help 499 Select this option to enable task and CPU time accounting on full 500 dynticks systems. This accounting is implemented by watching every 501 kernel-user boundaries using the context tracking subsystem. 502 The accounting is thus performed at the expense of some significant 503 overhead. 504 505 For now this is only useful if you are working on the full 506 dynticks subsystem development. 507 508 If unsure, say N. 509 510endchoice 511 512config IRQ_TIME_ACCOUNTING 513 bool "Fine granularity task level IRQ time accounting" 514 depends on HAVE_IRQ_TIME_ACCOUNTING && !VIRT_CPU_ACCOUNTING_NATIVE 515 help 516 Select this option to enable fine granularity task irq time 517 accounting. This is done by reading a timestamp on each 518 transitions between softirq and hardirq state, so there can be a 519 small performance impact. 520 521 If in doubt, say N here. 522 523config HAVE_SCHED_AVG_IRQ 524 def_bool y 525 depends on IRQ_TIME_ACCOUNTING || PARAVIRT_TIME_ACCOUNTING 526 depends on SMP 527 528config SCHED_THERMAL_PRESSURE 529 bool 530 default y if ARM && ARM_CPU_TOPOLOGY 531 default y if ARM64 532 depends on SMP 533 depends on CPU_FREQ_THERMAL 534 help 535 Select this option to enable thermal pressure accounting in the 536 scheduler. Thermal pressure is the value conveyed to the scheduler 537 that reflects the reduction in CPU compute capacity resulted from 538 thermal throttling. Thermal throttling occurs when the performance of 539 a CPU is capped due to high operating temperatures. 540 541 If selected, the scheduler will be able to balance tasks accordingly, 542 i.e. put less load on throttled CPUs than on non/less throttled ones. 543 544 This requires the architecture to implement 545 arch_update_thermal_pressure() and arch_scale_thermal_pressure(). 546 547config BSD_PROCESS_ACCT 548 bool "BSD Process Accounting" 549 depends on MULTIUSER 550 help 551 If you say Y here, a user level program will be able to instruct the 552 kernel (via a special system call) to write process accounting 553 information to a file: whenever a process exits, information about 554 that process will be appended to the file by the kernel. The 555 information includes things such as creation time, owning user, 556 command name, memory usage, controlling terminal etc. (the complete 557 list is in the struct acct in <file:include/linux/acct.h>). It is 558 up to the user level program to do useful things with this 559 information. This is generally a good idea, so say Y. 560 561config BSD_PROCESS_ACCT_V3 562 bool "BSD Process Accounting version 3 file format" 563 depends on BSD_PROCESS_ACCT 564 default n 565 help 566 If you say Y here, the process accounting information is written 567 in a new file format that also logs the process IDs of each 568 process and its parent. Note that this file format is incompatible 569 with previous v0/v1/v2 file formats, so you will need updated tools 570 for processing it. A preliminary version of these tools is available 571 at <http://www.gnu.org/software/acct/>. 572 573config TASKSTATS 574 bool "Export task/process statistics through netlink" 575 depends on NET 576 depends on MULTIUSER 577 default n 578 help 579 Export selected statistics for tasks/processes through the 580 generic netlink interface. Unlike BSD process accounting, the 581 statistics are available during the lifetime of tasks/processes as 582 responses to commands. Like BSD accounting, they are sent to user 583 space on task exit. 584 585 Say N if unsure. 586 587config TASK_DELAY_ACCT 588 bool "Enable per-task delay accounting" 589 depends on TASKSTATS 590 select SCHED_INFO 591 help 592 Collect information on time spent by a task waiting for system 593 resources like cpu, synchronous block I/O completion and swapping 594 in pages. Such statistics can help in setting a task's priorities 595 relative to other tasks for cpu, io, rss limits etc. 596 597 Say N if unsure. 598 599config TASK_XACCT 600 bool "Enable extended accounting over taskstats" 601 depends on TASKSTATS 602 help 603 Collect extended task accounting data and send the data 604 to userland for processing over the taskstats interface. 605 606 Say N if unsure. 607 608config TASK_IO_ACCOUNTING 609 bool "Enable per-task storage I/O accounting" 610 depends on TASK_XACCT 611 help 612 Collect information on the number of bytes of storage I/O which this 613 task has caused. 614 615 Say N if unsure. 616 617config PSI 618 bool "Pressure stall information tracking" 619 help 620 Collect metrics that indicate how overcommitted the CPU, memory, 621 and IO capacity are in the system. 622 623 If you say Y here, the kernel will create /proc/pressure/ with the 624 pressure statistics files cpu, memory, and io. These will indicate 625 the share of walltime in which some or all tasks in the system are 626 delayed due to contention of the respective resource. 627 628 In kernels with cgroup support, cgroups (cgroup2 only) will 629 have cpu.pressure, memory.pressure, and io.pressure files, 630 which aggregate pressure stalls for the grouped tasks only. 631 632 For more details see Documentation/accounting/psi.rst. 633 634 Say N if unsure. 635 636config PSI_DEFAULT_DISABLED 637 bool "Require boot parameter to enable pressure stall information tracking" 638 default n 639 depends on PSI 640 help 641 If set, pressure stall information tracking will be disabled 642 per default but can be enabled through passing psi=1 on the 643 kernel commandline during boot. 644 645 This feature adds some code to the task wakeup and sleep 646 paths of the scheduler. The overhead is too low to affect 647 common scheduling-intense workloads in practice (such as 648 webservers, memcache), but it does show up in artificial 649 scheduler stress tests, such as hackbench. 650 651 If you are paranoid and not sure what the kernel will be 652 used for, say Y. 653 654 Say N if unsure. 655 656endmenu # "CPU/Task time and stats accounting" 657 658config CPU_ISOLATION 659 bool "CPU isolation" 660 depends on SMP || COMPILE_TEST 661 default y 662 help 663 Make sure that CPUs running critical tasks are not disturbed by 664 any source of "noise" such as unbound workqueues, timers, kthreads... 665 Unbound jobs get offloaded to housekeeping CPUs. This is driven by 666 the "isolcpus=" boot parameter. 667 668 Say Y if unsure. 669 670source "kernel/rcu/Kconfig" 671 672config BUILD_BIN2C 673 bool 674 default n 675 676config IKCONFIG 677 tristate "Kernel .config support" 678 help 679 This option enables the complete Linux kernel ".config" file 680 contents to be saved in the kernel. It provides documentation 681 of which kernel options are used in a running kernel or in an 682 on-disk kernel. This information can be extracted from the kernel 683 image file with the script scripts/extract-ikconfig and used as 684 input to rebuild the current kernel or to build another kernel. 685 It can also be extracted from a running kernel by reading 686 /proc/config.gz if enabled (below). 687 688config IKCONFIG_PROC 689 bool "Enable access to .config through /proc/config.gz" 690 depends on IKCONFIG && PROC_FS 691 help 692 This option enables access to the kernel configuration file 693 through /proc/config.gz. 694 695config IKHEADERS 696 tristate "Enable kernel headers through /sys/kernel/kheaders.tar.xz" 697 depends on SYSFS 698 help 699 This option enables access to the in-kernel headers that are generated during 700 the build process. These can be used to build eBPF tracing programs, 701 or similar programs. If you build the headers as a module, a module called 702 kheaders.ko is built which can be loaded on-demand to get access to headers. 703 704config LOG_BUF_SHIFT 705 int "Kernel log buffer size (16 => 64KB, 17 => 128KB)" 706 range 12 25 707 default 17 708 depends on PRINTK 709 help 710 Select the minimal kernel log buffer size as a power of 2. 711 The final size is affected by LOG_CPU_MAX_BUF_SHIFT config 712 parameter, see below. Any higher size also might be forced 713 by "log_buf_len" boot parameter. 714 715 Examples: 716 17 => 128 KB 717 16 => 64 KB 718 15 => 32 KB 719 14 => 16 KB 720 13 => 8 KB 721 12 => 4 KB 722 723config LOG_CPU_MAX_BUF_SHIFT 724 int "CPU kernel log buffer size contribution (13 => 8 KB, 17 => 128KB)" 725 depends on SMP 726 range 0 21 727 default 12 if !BASE_SMALL 728 default 0 if BASE_SMALL 729 depends on PRINTK 730 help 731 This option allows to increase the default ring buffer size 732 according to the number of CPUs. The value defines the contribution 733 of each CPU as a power of 2. The used space is typically only few 734 lines however it might be much more when problems are reported, 735 e.g. backtraces. 736 737 The increased size means that a new buffer has to be allocated and 738 the original static one is unused. It makes sense only on systems 739 with more CPUs. Therefore this value is used only when the sum of 740 contributions is greater than the half of the default kernel ring 741 buffer as defined by LOG_BUF_SHIFT. The default values are set 742 so that more than 16 CPUs are needed to trigger the allocation. 743 744 Also this option is ignored when "log_buf_len" kernel parameter is 745 used as it forces an exact (power of two) size of the ring buffer. 746 747 The number of possible CPUs is used for this computation ignoring 748 hotplugging making the computation optimal for the worst case 749 scenario while allowing a simple algorithm to be used from bootup. 750 751 Examples shift values and their meaning: 752 17 => 128 KB for each CPU 753 16 => 64 KB for each CPU 754 15 => 32 KB for each CPU 755 14 => 16 KB for each CPU 756 13 => 8 KB for each CPU 757 12 => 4 KB for each CPU 758 759config PRINTK_SAFE_LOG_BUF_SHIFT 760 int "Temporary per-CPU printk log buffer size (12 => 4KB, 13 => 8KB)" 761 range 10 21 762 default 13 763 depends on PRINTK 764 help 765 Select the size of an alternate printk per-CPU buffer where messages 766 printed from usafe contexts are temporary stored. One example would 767 be NMI messages, another one - printk recursion. The messages are 768 copied to the main log buffer in a safe context to avoid a deadlock. 769 The value defines the size as a power of 2. 770 771 Those messages are rare and limited. The largest one is when 772 a backtrace is printed. It usually fits into 4KB. Select 773 8KB if you want to be on the safe side. 774 775 Examples: 776 17 => 128 KB for each CPU 777 16 => 64 KB for each CPU 778 15 => 32 KB for each CPU 779 14 => 16 KB for each CPU 780 13 => 8 KB for each CPU 781 12 => 4 KB for each CPU 782 783config PRINTK_INDEX 784 bool "Printk indexing debugfs interface" 785 depends on PRINTK && DEBUG_FS 786 help 787 Add support for indexing of all printk formats known at compile time 788 at <debugfs>/printk/index/<module>. 789 790 This can be used as part of maintaining daemons which monitor 791 /dev/kmsg, as it permits auditing the printk formats present in a 792 kernel, allowing detection of cases where monitored printks are 793 changed or no longer present. 794 795 There is no additional runtime cost to printk with this enabled. 796 797# 798# Architectures with an unreliable sched_clock() should select this: 799# 800config HAVE_UNSTABLE_SCHED_CLOCK 801 bool 802 803config GENERIC_SCHED_CLOCK 804 bool 805 806menu "Scheduler features" 807 808config UCLAMP_TASK 809 bool "Enable utilization clamping for RT/FAIR tasks" 810 depends on CPU_FREQ_GOV_SCHEDUTIL 811 help 812 This feature enables the scheduler to track the clamped utilization 813 of each CPU based on RUNNABLE tasks scheduled on that CPU. 814 815 With this option, the user can specify the min and max CPU 816 utilization allowed for RUNNABLE tasks. The max utilization defines 817 the maximum frequency a task should use while the min utilization 818 defines the minimum frequency it should use. 819 820 Both min and max utilization clamp values are hints to the scheduler, 821 aiming at improving its frequency selection policy, but they do not 822 enforce or grant any specific bandwidth for tasks. 823 824 If in doubt, say N. 825 826config UCLAMP_BUCKETS_COUNT 827 int "Number of supported utilization clamp buckets" 828 range 5 20 829 default 5 830 depends on UCLAMP_TASK 831 help 832 Defines the number of clamp buckets to use. The range of each bucket 833 will be SCHED_CAPACITY_SCALE/UCLAMP_BUCKETS_COUNT. The higher the 834 number of clamp buckets the finer their granularity and the higher 835 the precision of clamping aggregation and tracking at run-time. 836 837 For example, with the minimum configuration value we will have 5 838 clamp buckets tracking 20% utilization each. A 25% boosted tasks will 839 be refcounted in the [20..39]% bucket and will set the bucket clamp 840 effective value to 25%. 841 If a second 30% boosted task should be co-scheduled on the same CPU, 842 that task will be refcounted in the same bucket of the first task and 843 it will boost the bucket clamp effective value to 30%. 844 The clamp effective value of a bucket is reset to its nominal value 845 (20% in the example above) when there are no more tasks refcounted in 846 that bucket. 847 848 An additional boost/capping margin can be added to some tasks. In the 849 example above the 25% task will be boosted to 30% until it exits the 850 CPU. If that should be considered not acceptable on certain systems, 851 it's always possible to reduce the margin by increasing the number of 852 clamp buckets to trade off used memory for run-time tracking 853 precision. 854 855 If in doubt, use the default value. 856 857endmenu 858 859# 860# For architectures that want to enable the support for NUMA-affine scheduler 861# balancing logic: 862# 863config ARCH_SUPPORTS_NUMA_BALANCING 864 bool 865 866# 867# For architectures that prefer to flush all TLBs after a number of pages 868# are unmapped instead of sending one IPI per page to flush. The architecture 869# must provide guarantees on what happens if a clean TLB cache entry is 870# written after the unmap. Details are in mm/rmap.c near the check for 871# should_defer_flush. The architecture should also consider if the full flush 872# and the refill costs are offset by the savings of sending fewer IPIs. 873config ARCH_WANT_BATCHED_UNMAP_TLB_FLUSH 874 bool 875 876config CC_HAS_INT128 877 def_bool !$(cc-option,$(m64-flag) -D__SIZEOF_INT128__=0) && 64BIT 878 879config CC_IMPLICIT_FALLTHROUGH 880 string 881 default "-Wimplicit-fallthrough=5" if CC_IS_GCC && $(cc-option,-Wimplicit-fallthrough=5) 882 default "-Wimplicit-fallthrough" if CC_IS_CLANG && $(cc-option,-Wunreachable-code-fallthrough) 883 884# 885# For architectures that know their GCC __int128 support is sound 886# 887config ARCH_SUPPORTS_INT128 888 bool 889 890# For architectures that (ab)use NUMA to represent different memory regions 891# all cpu-local but of different latencies, such as SuperH. 892# 893config ARCH_WANT_NUMA_VARIABLE_LOCALITY 894 bool 895 896config NUMA_BALANCING 897 bool "Memory placement aware NUMA scheduler" 898 depends on ARCH_SUPPORTS_NUMA_BALANCING 899 depends on !ARCH_WANT_NUMA_VARIABLE_LOCALITY 900 depends on SMP && NUMA && MIGRATION && !PREEMPT_RT 901 help 902 This option adds support for automatic NUMA aware memory/task placement. 903 The mechanism is quite primitive and is based on migrating memory when 904 it has references to the node the task is running on. 905 906 This system will be inactive on UMA systems. 907 908config NUMA_BALANCING_DEFAULT_ENABLED 909 bool "Automatically enable NUMA aware memory/task placement" 910 default y 911 depends on NUMA_BALANCING 912 help 913 If set, automatic NUMA balancing will be enabled if running on a NUMA 914 machine. 915 916menuconfig CGROUPS 917 bool "Control Group support" 918 select KERNFS 919 help 920 This option adds support for grouping sets of processes together, for 921 use with process control subsystems such as Cpusets, CFS, memory 922 controls or device isolation. 923 See 924 - Documentation/scheduler/sched-design-CFS.rst (CFS) 925 - Documentation/admin-guide/cgroup-v1/ (features for grouping, isolation 926 and resource control) 927 928 Say N if unsure. 929 930if CGROUPS 931 932config PAGE_COUNTER 933 bool 934 935config MEMCG 936 bool "Memory controller" 937 select PAGE_COUNTER 938 select EVENTFD 939 help 940 Provides control over the memory footprint of tasks in a cgroup. 941 942config MEMCG_SWAP 943 bool 944 depends on MEMCG && SWAP 945 default y 946 947config MEMCG_KMEM 948 bool 949 depends on MEMCG && !SLOB 950 default y 951 952config BLK_CGROUP 953 bool "IO controller" 954 depends on BLOCK 955 default n 956 help 957 Generic block IO controller cgroup interface. This is the common 958 cgroup interface which should be used by various IO controlling 959 policies. 960 961 Currently, CFQ IO scheduler uses it to recognize task groups and 962 control disk bandwidth allocation (proportional time slice allocation) 963 to such task groups. It is also used by bio throttling logic in 964 block layer to implement upper limit in IO rates on a device. 965 966 This option only enables generic Block IO controller infrastructure. 967 One needs to also enable actual IO controlling logic/policy. For 968 enabling proportional weight division of disk bandwidth in CFQ, set 969 CONFIG_BFQ_GROUP_IOSCHED=y; for enabling throttling policy, set 970 CONFIG_BLK_DEV_THROTTLING=y. 971 972 See Documentation/admin-guide/cgroup-v1/blkio-controller.rst for more information. 973 974config CGROUP_WRITEBACK 975 bool 976 depends on MEMCG && BLK_CGROUP 977 default y 978 979menuconfig CGROUP_SCHED 980 bool "CPU controller" 981 default n 982 help 983 This feature lets CPU scheduler recognize task groups and control CPU 984 bandwidth allocation to such task groups. It uses cgroups to group 985 tasks. 986 987if CGROUP_SCHED 988config FAIR_GROUP_SCHED 989 bool "Group scheduling for SCHED_OTHER" 990 depends on CGROUP_SCHED 991 default CGROUP_SCHED 992 993config CFS_BANDWIDTH 994 bool "CPU bandwidth provisioning for FAIR_GROUP_SCHED" 995 depends on FAIR_GROUP_SCHED 996 default n 997 help 998 This option allows users to define CPU bandwidth rates (limits) for 999 tasks running within the fair group scheduler. Groups with no limit 1000 set are considered to be unconstrained and will run with no 1001 restriction. 1002 See Documentation/scheduler/sched-bwc.rst for more information. 1003 1004config RT_GROUP_SCHED 1005 bool "Group scheduling for SCHED_RR/FIFO" 1006 depends on CGROUP_SCHED 1007 default n 1008 help 1009 This feature lets you explicitly allocate real CPU bandwidth 1010 to task groups. If enabled, it will also make it impossible to 1011 schedule realtime tasks for non-root users until you allocate 1012 realtime bandwidth for them. 1013 See Documentation/scheduler/sched-rt-group.rst for more information. 1014 1015endif #CGROUP_SCHED 1016 1017config UCLAMP_TASK_GROUP 1018 bool "Utilization clamping per group of tasks" 1019 depends on CGROUP_SCHED 1020 depends on UCLAMP_TASK 1021 default n 1022 help 1023 This feature enables the scheduler to track the clamped utilization 1024 of each CPU based on RUNNABLE tasks currently scheduled on that CPU. 1025 1026 When this option is enabled, the user can specify a min and max 1027 CPU bandwidth which is allowed for each single task in a group. 1028 The max bandwidth allows to clamp the maximum frequency a task 1029 can use, while the min bandwidth allows to define a minimum 1030 frequency a task will always use. 1031 1032 When task group based utilization clamping is enabled, an eventually 1033 specified task-specific clamp value is constrained by the cgroup 1034 specified clamp value. Both minimum and maximum task clamping cannot 1035 be bigger than the corresponding clamping defined at task group level. 1036 1037 If in doubt, say N. 1038 1039config CGROUP_PIDS 1040 bool "PIDs controller" 1041 help 1042 Provides enforcement of process number limits in the scope of a 1043 cgroup. Any attempt to fork more processes than is allowed in the 1044 cgroup will fail. PIDs are fundamentally a global resource because it 1045 is fairly trivial to reach PID exhaustion before you reach even a 1046 conservative kmemcg limit. As a result, it is possible to grind a 1047 system to halt without being limited by other cgroup policies. The 1048 PIDs controller is designed to stop this from happening. 1049 1050 It should be noted that organisational operations (such as attaching 1051 to a cgroup hierarchy) will *not* be blocked by the PIDs controller, 1052 since the PIDs limit only affects a process's ability to fork, not to 1053 attach to a cgroup. 1054 1055config CGROUP_RDMA 1056 bool "RDMA controller" 1057 help 1058 Provides enforcement of RDMA resources defined by IB stack. 1059 It is fairly easy for consumers to exhaust RDMA resources, which 1060 can result into resource unavailability to other consumers. 1061 RDMA controller is designed to stop this from happening. 1062 Attaching processes with active RDMA resources to the cgroup 1063 hierarchy is allowed even if can cross the hierarchy's limit. 1064 1065config CGROUP_FREEZER 1066 bool "Freezer controller" 1067 help 1068 Provides a way to freeze and unfreeze all tasks in a 1069 cgroup. 1070 1071 This option affects the ORIGINAL cgroup interface. The cgroup2 memory 1072 controller includes important in-kernel memory consumers per default. 1073 1074 If you're using cgroup2, say N. 1075 1076config CGROUP_HUGETLB 1077 bool "HugeTLB controller" 1078 depends on HUGETLB_PAGE 1079 select PAGE_COUNTER 1080 default n 1081 help 1082 Provides a cgroup controller for HugeTLB pages. 1083 When you enable this, you can put a per cgroup limit on HugeTLB usage. 1084 The limit is enforced during page fault. Since HugeTLB doesn't 1085 support page reclaim, enforcing the limit at page fault time implies 1086 that, the application will get SIGBUS signal if it tries to access 1087 HugeTLB pages beyond its limit. This requires the application to know 1088 beforehand how much HugeTLB pages it would require for its use. The 1089 control group is tracked in the third page lru pointer. This means 1090 that we cannot use the controller with huge page less than 3 pages. 1091 1092config CPUSETS 1093 bool "Cpuset controller" 1094 depends on SMP 1095 help 1096 This option will let you create and manage CPUSETs which 1097 allow dynamically partitioning a system into sets of CPUs and 1098 Memory Nodes and assigning tasks to run only within those sets. 1099 This is primarily useful on large SMP or NUMA systems. 1100 1101 Say N if unsure. 1102 1103config PROC_PID_CPUSET 1104 bool "Include legacy /proc/<pid>/cpuset file" 1105 depends on CPUSETS 1106 default y 1107 1108config CGROUP_DEVICE 1109 bool "Device controller" 1110 help 1111 Provides a cgroup controller implementing whitelists for 1112 devices which a process in the cgroup can mknod or open. 1113 1114config CGROUP_CPUACCT 1115 bool "Simple CPU accounting controller" 1116 help 1117 Provides a simple controller for monitoring the 1118 total CPU consumed by the tasks in a cgroup. 1119 1120config CGROUP_PERF 1121 bool "Perf controller" 1122 depends on PERF_EVENTS 1123 help 1124 This option extends the perf per-cpu mode to restrict monitoring 1125 to threads which belong to the cgroup specified and run on the 1126 designated cpu. Or this can be used to have cgroup ID in samples 1127 so that it can monitor performance events among cgroups. 1128 1129 Say N if unsure. 1130 1131config CGROUP_BPF 1132 bool "Support for eBPF programs attached to cgroups" 1133 depends on BPF_SYSCALL 1134 select SOCK_CGROUP_DATA 1135 help 1136 Allow attaching eBPF programs to a cgroup using the bpf(2) 1137 syscall command BPF_PROG_ATTACH. 1138 1139 In which context these programs are accessed depends on the type 1140 of attachment. For instance, programs that are attached using 1141 BPF_CGROUP_INET_INGRESS will be executed on the ingress path of 1142 inet sockets. 1143 1144config CGROUP_MISC 1145 bool "Misc resource controller" 1146 default n 1147 help 1148 Provides a controller for miscellaneous resources on a host. 1149 1150 Miscellaneous scalar resources are the resources on the host system 1151 which cannot be abstracted like the other cgroups. This controller 1152 tracks and limits the miscellaneous resources used by a process 1153 attached to a cgroup hierarchy. 1154 1155 For more information, please check misc cgroup section in 1156 /Documentation/admin-guide/cgroup-v2.rst. 1157 1158config CGROUP_DEBUG 1159 bool "Debug controller" 1160 default n 1161 depends on DEBUG_KERNEL 1162 help 1163 This option enables a simple controller that exports 1164 debugging information about the cgroups framework. This 1165 controller is for control cgroup debugging only. Its 1166 interfaces are not stable. 1167 1168 Say N. 1169 1170config SOCK_CGROUP_DATA 1171 bool 1172 default n 1173 1174endif # CGROUPS 1175 1176menuconfig NAMESPACES 1177 bool "Namespaces support" if EXPERT 1178 depends on MULTIUSER 1179 default !EXPERT 1180 help 1181 Provides the way to make tasks work with different objects using 1182 the same id. For example same IPC id may refer to different objects 1183 or same user id or pid may refer to different tasks when used in 1184 different namespaces. 1185 1186if NAMESPACES 1187 1188config UTS_NS 1189 bool "UTS namespace" 1190 default y 1191 help 1192 In this namespace tasks see different info provided with the 1193 uname() system call 1194 1195config TIME_NS 1196 bool "TIME namespace" 1197 depends on GENERIC_VDSO_TIME_NS 1198 default y 1199 help 1200 In this namespace boottime and monotonic clocks can be set. 1201 The time will keep going with the same pace. 1202 1203config IPC_NS 1204 bool "IPC namespace" 1205 depends on (SYSVIPC || POSIX_MQUEUE) 1206 default y 1207 help 1208 In this namespace tasks work with IPC ids which correspond to 1209 different IPC objects in different namespaces. 1210 1211config USER_NS 1212 bool "User namespace" 1213 default n 1214 help 1215 This allows containers, i.e. vservers, to use user namespaces 1216 to provide different user info for different servers. 1217 1218 When user namespaces are enabled in the kernel it is 1219 recommended that the MEMCG option also be enabled and that 1220 user-space use the memory control groups to limit the amount 1221 of memory a memory unprivileged users can use. 1222 1223 If unsure, say N. 1224 1225config PID_NS 1226 bool "PID Namespaces" 1227 default y 1228 help 1229 Support process id namespaces. This allows having multiple 1230 processes with the same pid as long as they are in different 1231 pid namespaces. This is a building block of containers. 1232 1233config NET_NS 1234 bool "Network namespace" 1235 depends on NET 1236 default y 1237 help 1238 Allow user space to create what appear to be multiple instances 1239 of the network stack. 1240 1241endif # NAMESPACES 1242 1243config CHECKPOINT_RESTORE 1244 bool "Checkpoint/restore support" 1245 select PROC_CHILDREN 1246 select KCMP 1247 default n 1248 help 1249 Enables additional kernel features in a sake of checkpoint/restore. 1250 In particular it adds auxiliary prctl codes to setup process text, 1251 data and heap segment sizes, and a few additional /proc filesystem 1252 entries. 1253 1254 If unsure, say N here. 1255 1256config SCHED_AUTOGROUP 1257 bool "Automatic process group scheduling" 1258 select CGROUPS 1259 select CGROUP_SCHED 1260 select FAIR_GROUP_SCHED 1261 help 1262 This option optimizes the scheduler for common desktop workloads by 1263 automatically creating and populating task groups. This separation 1264 of workloads isolates aggressive CPU burners (like build jobs) from 1265 desktop applications. Task group autogeneration is currently based 1266 upon task session. 1267 1268config SYSFS_DEPRECATED 1269 bool "Enable deprecated sysfs features to support old userspace tools" 1270 depends on SYSFS 1271 default n 1272 help 1273 This option adds code that switches the layout of the "block" class 1274 devices, to not show up in /sys/class/block/, but only in 1275 /sys/block/. 1276 1277 This switch is only active when the sysfs.deprecated=1 boot option is 1278 passed or the SYSFS_DEPRECATED_V2 option is set. 1279 1280 This option allows new kernels to run on old distributions and tools, 1281 which might get confused by /sys/class/block/. Since 2007/2008 all 1282 major distributions and tools handle this just fine. 1283 1284 Recent distributions and userspace tools after 2009/2010 depend on 1285 the existence of /sys/class/block/, and will not work with this 1286 option enabled. 1287 1288 Only if you are using a new kernel on an old distribution, you might 1289 need to say Y here. 1290 1291config SYSFS_DEPRECATED_V2 1292 bool "Enable deprecated sysfs features by default" 1293 default n 1294 depends on SYSFS 1295 depends on SYSFS_DEPRECATED 1296 help 1297 Enable deprecated sysfs by default. 1298 1299 See the CONFIG_SYSFS_DEPRECATED option for more details about this 1300 option. 1301 1302 Only if you are using a new kernel on an old distribution, you might 1303 need to say Y here. Even then, odds are you would not need it 1304 enabled, you can always pass the boot option if absolutely necessary. 1305 1306config RELAY 1307 bool "Kernel->user space relay support (formerly relayfs)" 1308 select IRQ_WORK 1309 help 1310 This option enables support for relay interface support in 1311 certain file systems (such as debugfs). 1312 It is designed to provide an efficient mechanism for tools and 1313 facilities to relay large amounts of data from kernel space to 1314 user space. 1315 1316 If unsure, say N. 1317 1318config BLK_DEV_INITRD 1319 bool "Initial RAM filesystem and RAM disk (initramfs/initrd) support" 1320 help 1321 The initial RAM filesystem is a ramfs which is loaded by the 1322 boot loader (loadlin or lilo) and that is mounted as root 1323 before the normal boot procedure. It is typically used to 1324 load modules needed to mount the "real" root file system, 1325 etc. See <file:Documentation/admin-guide/initrd.rst> for details. 1326 1327 If RAM disk support (BLK_DEV_RAM) is also included, this 1328 also enables initial RAM disk (initrd) support and adds 1329 15 Kbytes (more on some other architectures) to the kernel size. 1330 1331 If unsure say Y. 1332 1333if BLK_DEV_INITRD 1334 1335source "usr/Kconfig" 1336 1337endif 1338 1339config BOOT_CONFIG 1340 bool "Boot config support" 1341 select BLK_DEV_INITRD if !BOOT_CONFIG_EMBED 1342 help 1343 Extra boot config allows system admin to pass a config file as 1344 complemental extension of kernel cmdline when booting. 1345 The boot config file must be attached at the end of initramfs 1346 with checksum, size and magic word. 1347 See <file:Documentation/admin-guide/bootconfig.rst> for details. 1348 1349 If unsure, say Y. 1350 1351config BOOT_CONFIG_EMBED 1352 bool "Embed bootconfig file in the kernel" 1353 depends on BOOT_CONFIG 1354 help 1355 Embed a bootconfig file given by BOOT_CONFIG_EMBED_FILE in the 1356 kernel. Usually, the bootconfig file is loaded with the initrd 1357 image. But if the system doesn't support initrd, this option will 1358 help you by embedding a bootconfig file while building the kernel. 1359 1360 If unsure, say N. 1361 1362config BOOT_CONFIG_EMBED_FILE 1363 string "Embedded bootconfig file path" 1364 depends on BOOT_CONFIG_EMBED 1365 help 1366 Specify a bootconfig file which will be embedded to the kernel. 1367 This bootconfig will be used if there is no initrd or no other 1368 bootconfig in the initrd. 1369 1370config INITRAMFS_PRESERVE_MTIME 1371 bool "Preserve cpio archive mtimes in initramfs" 1372 default y 1373 help 1374 Each entry in an initramfs cpio archive carries an mtime value. When 1375 enabled, extracted cpio items take this mtime, with directory mtime 1376 setting deferred until after creation of any child entries. 1377 1378 If unsure, say Y. 1379 1380choice 1381 prompt "Compiler optimization level" 1382 default CC_OPTIMIZE_FOR_PERFORMANCE 1383 1384config CC_OPTIMIZE_FOR_PERFORMANCE 1385 bool "Optimize for performance (-O2)" 1386 help 1387 This is the default optimization level for the kernel, building 1388 with the "-O2" compiler flag for best performance and most 1389 helpful compile-time warnings. 1390 1391config CC_OPTIMIZE_FOR_PERFORMANCE_O3 1392 bool "Optimize more for performance (-O3)" 1393 depends on ARC 1394 help 1395 Choosing this option will pass "-O3" to your compiler to optimize 1396 the kernel yet more for performance. 1397 1398config CC_OPTIMIZE_FOR_SIZE 1399 bool "Optimize for size (-Os)" 1400 help 1401 Choosing this option will pass "-Os" to your compiler resulting 1402 in a smaller kernel. 1403 1404endchoice 1405 1406config HAVE_LD_DEAD_CODE_DATA_ELIMINATION 1407 bool 1408 help 1409 This requires that the arch annotates or otherwise protects 1410 its external entry points from being discarded. Linker scripts 1411 must also merge .text.*, .data.*, and .bss.* correctly into 1412 output sections. Care must be taken not to pull in unrelated 1413 sections (e.g., '.text.init'). Typically '.' in section names 1414 is used to distinguish them from label names / C identifiers. 1415 1416config LD_DEAD_CODE_DATA_ELIMINATION 1417 bool "Dead code and data elimination (EXPERIMENTAL)" 1418 depends on HAVE_LD_DEAD_CODE_DATA_ELIMINATION 1419 depends on EXPERT 1420 depends on $(cc-option,-ffunction-sections -fdata-sections) 1421 depends on $(ld-option,--gc-sections) 1422 help 1423 Enable this if you want to do dead code and data elimination with 1424 the linker by compiling with -ffunction-sections -fdata-sections, 1425 and linking with --gc-sections. 1426 1427 This can reduce on disk and in-memory size of the kernel 1428 code and static data, particularly for small configs and 1429 on small systems. This has the possibility of introducing 1430 silently broken kernel if the required annotations are not 1431 present. This option is not well tested yet, so use at your 1432 own risk. 1433 1434config LD_ORPHAN_WARN 1435 def_bool y 1436 depends on ARCH_WANT_LD_ORPHAN_WARN 1437 depends on $(ld-option,--orphan-handling=warn) 1438 1439config SYSCTL 1440 bool 1441 1442config HAVE_UID16 1443 bool 1444 1445config SYSCTL_EXCEPTION_TRACE 1446 bool 1447 help 1448 Enable support for /proc/sys/debug/exception-trace. 1449 1450config SYSCTL_ARCH_UNALIGN_NO_WARN 1451 bool 1452 help 1453 Enable support for /proc/sys/kernel/ignore-unaligned-usertrap 1454 Allows arch to define/use @no_unaligned_warning to possibly warn 1455 about unaligned access emulation going on under the hood. 1456 1457config SYSCTL_ARCH_UNALIGN_ALLOW 1458 bool 1459 help 1460 Enable support for /proc/sys/kernel/unaligned-trap 1461 Allows arches to define/use @unaligned_enabled to runtime toggle 1462 the unaligned access emulation. 1463 see arch/parisc/kernel/unaligned.c for reference 1464 1465config HAVE_PCSPKR_PLATFORM 1466 bool 1467 1468# interpreter that classic socket filters depend on 1469config BPF 1470 bool 1471 1472menuconfig EXPERT 1473 bool "Configure standard kernel features (expert users)" 1474 # Unhide debug options, to make the on-by-default options visible 1475 select DEBUG_KERNEL 1476 help 1477 This option allows certain base kernel options and settings 1478 to be disabled or tweaked. This is for specialized 1479 environments which can tolerate a "non-standard" kernel. 1480 Only use this if you really know what you are doing. 1481 1482config UID16 1483 bool "Enable 16-bit UID system calls" if EXPERT 1484 depends on HAVE_UID16 && MULTIUSER 1485 default y 1486 help 1487 This enables the legacy 16-bit UID syscall wrappers. 1488 1489config MULTIUSER 1490 bool "Multiple users, groups and capabilities support" if EXPERT 1491 default y 1492 help 1493 This option enables support for non-root users, groups and 1494 capabilities. 1495 1496 If you say N here, all processes will run with UID 0, GID 0, and all 1497 possible capabilities. Saying N here also compiles out support for 1498 system calls related to UIDs, GIDs, and capabilities, such as setuid, 1499 setgid, and capset. 1500 1501 If unsure, say Y here. 1502 1503config SGETMASK_SYSCALL 1504 bool "sgetmask/ssetmask syscalls support" if EXPERT 1505 def_bool PARISC || M68K || PPC || MIPS || X86 || SPARC || MICROBLAZE || SUPERH 1506 help 1507 sys_sgetmask and sys_ssetmask are obsolete system calls 1508 no longer supported in libc but still enabled by default in some 1509 architectures. 1510 1511 If unsure, leave the default option here. 1512 1513config SYSFS_SYSCALL 1514 bool "Sysfs syscall support" if EXPERT 1515 default y 1516 help 1517 sys_sysfs is an obsolete system call no longer supported in libc. 1518 Note that disabling this option is more secure but might break 1519 compatibility with some systems. 1520 1521 If unsure say Y here. 1522 1523config FHANDLE 1524 bool "open by fhandle syscalls" if EXPERT 1525 select EXPORTFS 1526 default y 1527 help 1528 If you say Y here, a user level program will be able to map 1529 file names to handle and then later use the handle for 1530 different file system operations. This is useful in implementing 1531 userspace file servers, which now track files using handles instead 1532 of names. The handle would remain the same even if file names 1533 get renamed. Enables open_by_handle_at(2) and name_to_handle_at(2) 1534 syscalls. 1535 1536config POSIX_TIMERS 1537 bool "Posix Clocks & timers" if EXPERT 1538 default y 1539 help 1540 This includes native support for POSIX timers to the kernel. 1541 Some embedded systems have no use for them and therefore they 1542 can be configured out to reduce the size of the kernel image. 1543 1544 When this option is disabled, the following syscalls won't be 1545 available: timer_create, timer_gettime: timer_getoverrun, 1546 timer_settime, timer_delete, clock_adjtime, getitimer, 1547 setitimer, alarm. Furthermore, the clock_settime, clock_gettime, 1548 clock_getres and clock_nanosleep syscalls will be limited to 1549 CLOCK_REALTIME, CLOCK_MONOTONIC and CLOCK_BOOTTIME only. 1550 1551 If unsure say y. 1552 1553config PRINTK 1554 default y 1555 bool "Enable support for printk" if EXPERT 1556 select IRQ_WORK 1557 help 1558 This option enables normal printk support. Removing it 1559 eliminates most of the message strings from the kernel image 1560 and makes the kernel more or less silent. As this makes it 1561 very difficult to diagnose system problems, saying N here is 1562 strongly discouraged. 1563 1564config BUG 1565 bool "BUG() support" if EXPERT 1566 default y 1567 help 1568 Disabling this option eliminates support for BUG and WARN, reducing 1569 the size of your kernel image and potentially quietly ignoring 1570 numerous fatal conditions. You should only consider disabling this 1571 option for embedded systems with no facilities for reporting errors. 1572 Just say Y. 1573 1574config ELF_CORE 1575 depends on COREDUMP 1576 default y 1577 bool "Enable ELF core dumps" if EXPERT 1578 help 1579 Enable support for generating core dumps. Disabling saves about 4k. 1580 1581 1582config PCSPKR_PLATFORM 1583 bool "Enable PC-Speaker support" if EXPERT 1584 depends on HAVE_PCSPKR_PLATFORM 1585 select I8253_LOCK 1586 default y 1587 help 1588 This option allows to disable the internal PC-Speaker 1589 support, saving some memory. 1590 1591config BASE_FULL 1592 default y 1593 bool "Enable full-sized data structures for core" if EXPERT 1594 help 1595 Disabling this option reduces the size of miscellaneous core 1596 kernel data structures. This saves memory on small machines, 1597 but may reduce performance. 1598 1599config FUTEX 1600 bool "Enable futex support" if EXPERT 1601 depends on !(SPARC32 && SMP) 1602 default y 1603 imply RT_MUTEXES 1604 help 1605 Disabling this option will cause the kernel to be built without 1606 support for "fast userspace mutexes". The resulting kernel may not 1607 run glibc-based applications correctly. 1608 1609config FUTEX_PI 1610 bool 1611 depends on FUTEX && RT_MUTEXES 1612 default y 1613 1614config EPOLL 1615 bool "Enable eventpoll support" if EXPERT 1616 default y 1617 help 1618 Disabling this option will cause the kernel to be built without 1619 support for epoll family of system calls. 1620 1621config SIGNALFD 1622 bool "Enable signalfd() system call" if EXPERT 1623 default y 1624 help 1625 Enable the signalfd() system call that allows to receive signals 1626 on a file descriptor. 1627 1628 If unsure, say Y. 1629 1630config TIMERFD 1631 bool "Enable timerfd() system call" if EXPERT 1632 default y 1633 help 1634 Enable the timerfd() system call that allows to receive timer 1635 events on a file descriptor. 1636 1637 If unsure, say Y. 1638 1639config EVENTFD 1640 bool "Enable eventfd() system call" if EXPERT 1641 default y 1642 help 1643 Enable the eventfd() system call that allows to receive both 1644 kernel notification (ie. KAIO) or userspace notifications. 1645 1646 If unsure, say Y. 1647 1648config SHMEM 1649 bool "Use full shmem filesystem" if EXPERT 1650 default y 1651 depends on MMU 1652 help 1653 The shmem is an internal filesystem used to manage shared memory. 1654 It is backed by swap and manages resource limits. It is also exported 1655 to userspace as tmpfs if TMPFS is enabled. Disabling this 1656 option replaces shmem and tmpfs with the much simpler ramfs code, 1657 which may be appropriate on small systems without swap. 1658 1659config AIO 1660 bool "Enable AIO support" if EXPERT 1661 default y 1662 help 1663 This option enables POSIX asynchronous I/O which may by used 1664 by some high performance threaded applications. Disabling 1665 this option saves about 7k. 1666 1667config IO_URING 1668 bool "Enable IO uring support" if EXPERT 1669 select IO_WQ 1670 default y 1671 help 1672 This option enables support for the io_uring interface, enabling 1673 applications to submit and complete IO through submission and 1674 completion rings that are shared between the kernel and application. 1675 1676config ADVISE_SYSCALLS 1677 bool "Enable madvise/fadvise syscalls" if EXPERT 1678 default y 1679 help 1680 This option enables the madvise and fadvise syscalls, used by 1681 applications to advise the kernel about their future memory or file 1682 usage, improving performance. If building an embedded system where no 1683 applications use these syscalls, you can disable this option to save 1684 space. 1685 1686config MEMBARRIER 1687 bool "Enable membarrier() system call" if EXPERT 1688 default y 1689 help 1690 Enable the membarrier() system call that allows issuing memory 1691 barriers across all running threads, which can be used to distribute 1692 the cost of user-space memory barriers asymmetrically by transforming 1693 pairs of memory barriers into pairs consisting of membarrier() and a 1694 compiler barrier. 1695 1696 If unsure, say Y. 1697 1698config KALLSYMS 1699 bool "Load all symbols for debugging/ksymoops" if EXPERT 1700 default y 1701 help 1702 Say Y here to let the kernel print out symbolic crash information and 1703 symbolic stack backtraces. This increases the size of the kernel 1704 somewhat, as all symbols have to be loaded into the kernel image. 1705 1706config KALLSYMS_ALL 1707 bool "Include all symbols in kallsyms" 1708 depends on DEBUG_KERNEL && KALLSYMS 1709 help 1710 Normally kallsyms only contains the symbols of functions for nicer 1711 OOPS messages and backtraces (i.e., symbols from the text and inittext 1712 sections). This is sufficient for most cases. And only in very rare 1713 cases (e.g., when a debugger is used) all symbols are required (e.g., 1714 names of variables from the data sections, etc). 1715 1716 This option makes sure that all symbols are loaded into the kernel 1717 image (i.e., symbols from all sections) in cost of increased kernel 1718 size (depending on the kernel configuration, it may be 300KiB or 1719 something like this). 1720 1721 Say N unless you really need all symbols. 1722 1723config KALLSYMS_ABSOLUTE_PERCPU 1724 bool 1725 depends on KALLSYMS 1726 default X86_64 && SMP 1727 1728config KALLSYMS_BASE_RELATIVE 1729 bool 1730 depends on KALLSYMS 1731 default !IA64 1732 help 1733 Instead of emitting them as absolute values in the native word size, 1734 emit the symbol references in the kallsyms table as 32-bit entries, 1735 each containing a relative value in the range [base, base + U32_MAX] 1736 or, when KALLSYMS_ABSOLUTE_PERCPU is in effect, each containing either 1737 an absolute value in the range [0, S32_MAX] or a relative value in the 1738 range [base, base + S32_MAX], where base is the lowest relative symbol 1739 address encountered in the image. 1740 1741 On 64-bit builds, this reduces the size of the address table by 50%, 1742 but more importantly, it results in entries whose values are build 1743 time constants, and no relocation pass is required at runtime to fix 1744 up the entries based on the runtime load address of the kernel. 1745 1746# end of the "standard kernel features (expert users)" menu 1747 1748# syscall, maps, verifier 1749 1750config ARCH_HAS_MEMBARRIER_CALLBACKS 1751 bool 1752 1753config ARCH_HAS_MEMBARRIER_SYNC_CORE 1754 bool 1755 1756config KCMP 1757 bool "Enable kcmp() system call" if EXPERT 1758 help 1759 Enable the kernel resource comparison system call. It provides 1760 user-space with the ability to compare two processes to see if they 1761 share a common resource, such as a file descriptor or even virtual 1762 memory space. 1763 1764 If unsure, say N. 1765 1766config RSEQ 1767 bool "Enable rseq() system call" if EXPERT 1768 default y 1769 depends on HAVE_RSEQ 1770 select MEMBARRIER 1771 help 1772 Enable the restartable sequences system call. It provides a 1773 user-space cache for the current CPU number value, which 1774 speeds up getting the current CPU number from user-space, 1775 as well as an ABI to speed up user-space operations on 1776 per-CPU data. 1777 1778 If unsure, say Y. 1779 1780config DEBUG_RSEQ 1781 default n 1782 bool "Enabled debugging of rseq() system call" if EXPERT 1783 depends on RSEQ && DEBUG_KERNEL 1784 help 1785 Enable extra debugging checks for the rseq system call. 1786 1787 If unsure, say N. 1788 1789config EMBEDDED 1790 bool "Embedded system" 1791 select EXPERT 1792 help 1793 This option should be enabled if compiling the kernel for 1794 an embedded system so certain expert options are available 1795 for configuration. 1796 1797config HAVE_PERF_EVENTS 1798 bool 1799 help 1800 See tools/perf/design.txt for details. 1801 1802config GUEST_PERF_EVENTS 1803 bool 1804 depends on HAVE_PERF_EVENTS 1805 1806config PERF_USE_VMALLOC 1807 bool 1808 help 1809 See tools/perf/design.txt for details 1810 1811config PC104 1812 bool "PC/104 support" if EXPERT 1813 help 1814 Expose PC/104 form factor device drivers and options available for 1815 selection and configuration. Enable this option if your target 1816 machine has a PC/104 bus. 1817 1818menu "Kernel Performance Events And Counters" 1819 1820config PERF_EVENTS 1821 bool "Kernel performance events and counters" 1822 default y if PROFILING 1823 depends on HAVE_PERF_EVENTS 1824 select IRQ_WORK 1825 select SRCU 1826 help 1827 Enable kernel support for various performance events provided 1828 by software and hardware. 1829 1830 Software events are supported either built-in or via the 1831 use of generic tracepoints. 1832 1833 Most modern CPUs support performance events via performance 1834 counter registers. These registers count the number of certain 1835 types of hw events: such as instructions executed, cachemisses 1836 suffered, or branches mis-predicted - without slowing down the 1837 kernel or applications. These registers can also trigger interrupts 1838 when a threshold number of events have passed - and can thus be 1839 used to profile the code that runs on that CPU. 1840 1841 The Linux Performance Event subsystem provides an abstraction of 1842 these software and hardware event capabilities, available via a 1843 system call and used by the "perf" utility in tools/perf/. It 1844 provides per task and per CPU counters, and it provides event 1845 capabilities on top of those. 1846 1847 Say Y if unsure. 1848 1849config DEBUG_PERF_USE_VMALLOC 1850 default n 1851 bool "Debug: use vmalloc to back perf mmap() buffers" 1852 depends on PERF_EVENTS && DEBUG_KERNEL && !PPC 1853 select PERF_USE_VMALLOC 1854 help 1855 Use vmalloc memory to back perf mmap() buffers. 1856 1857 Mostly useful for debugging the vmalloc code on platforms 1858 that don't require it. 1859 1860 Say N if unsure. 1861 1862endmenu 1863 1864config SYSTEM_DATA_VERIFICATION 1865 def_bool n 1866 select SYSTEM_TRUSTED_KEYRING 1867 select KEYS 1868 select CRYPTO 1869 select CRYPTO_RSA 1870 select ASYMMETRIC_KEY_TYPE 1871 select ASYMMETRIC_PUBLIC_KEY_SUBTYPE 1872 select ASN1 1873 select OID_REGISTRY 1874 select X509_CERTIFICATE_PARSER 1875 select PKCS7_MESSAGE_PARSER 1876 help 1877 Provide PKCS#7 message verification using the contents of the system 1878 trusted keyring to provide public keys. This then can be used for 1879 module verification, kexec image verification and firmware blob 1880 verification. 1881 1882config PROFILING 1883 bool "Profiling support" 1884 help 1885 Say Y here to enable the extended profiling support mechanisms used 1886 by profilers. 1887 1888# 1889# Place an empty function call at each tracepoint site. Can be 1890# dynamically changed for a probe function. 1891# 1892config TRACEPOINTS 1893 bool 1894 1895endmenu # General setup 1896 1897source "arch/Kconfig" 1898 1899config RT_MUTEXES 1900 bool 1901 default y if PREEMPT_RT 1902 1903config BASE_SMALL 1904 int 1905 default 0 if BASE_FULL 1906 default 1 if !BASE_FULL 1907 1908config MODULE_SIG_FORMAT 1909 def_bool n 1910 select SYSTEM_DATA_VERIFICATION 1911 1912menuconfig MODULES 1913 bool "Enable loadable module support" 1914 modules 1915 help 1916 Kernel modules are small pieces of compiled code which can 1917 be inserted in the running kernel, rather than being 1918 permanently built into the kernel. You use the "modprobe" 1919 tool to add (and sometimes remove) them. If you say Y here, 1920 many parts of the kernel can be built as modules (by 1921 answering M instead of Y where indicated): this is most 1922 useful for infrequently used options which are not required 1923 for booting. For more information, see the man pages for 1924 modprobe, lsmod, modinfo, insmod and rmmod. 1925 1926 If you say Y here, you will need to run "make 1927 modules_install" to put the modules under /lib/modules/ 1928 where modprobe can find them (you may need to be root to do 1929 this). 1930 1931 If unsure, say Y. 1932 1933if MODULES 1934 1935config MODULE_FORCE_LOAD 1936 bool "Forced module loading" 1937 default n 1938 help 1939 Allow loading of modules without version information (ie. modprobe 1940 --force). Forced module loading sets the 'F' (forced) taint flag and 1941 is usually a really bad idea. 1942 1943config MODULE_UNLOAD 1944 bool "Module unloading" 1945 help 1946 Without this option you will not be able to unload any 1947 modules (note that some modules may not be unloadable 1948 anyway), which makes your kernel smaller, faster 1949 and simpler. If unsure, say Y. 1950 1951config MODULE_FORCE_UNLOAD 1952 bool "Forced module unloading" 1953 depends on MODULE_UNLOAD 1954 help 1955 This option allows you to force a module to unload, even if the 1956 kernel believes it is unsafe: the kernel will remove the module 1957 without waiting for anyone to stop using it (using the -f option to 1958 rmmod). This is mainly for kernel developers and desperate users. 1959 If unsure, say N. 1960 1961config MODULE_UNLOAD_TAINT_TRACKING 1962 bool "Tainted module unload tracking" 1963 depends on MODULE_UNLOAD 1964 default n 1965 help 1966 This option allows you to maintain a record of each unloaded 1967 module that tainted the kernel. In addition to displaying a 1968 list of linked (or loaded) modules e.g. on detection of a bad 1969 page (see bad_page()), the aforementioned details are also 1970 shown. If unsure, say N. 1971 1972config MODVERSIONS 1973 bool "Module versioning support" 1974 help 1975 Usually, you have to use modules compiled with your kernel. 1976 Saying Y here makes it sometimes possible to use modules 1977 compiled for different kernels, by adding enough information 1978 to the modules to (hopefully) spot any changes which would 1979 make them incompatible with the kernel you are running. If 1980 unsure, say N. 1981 1982config ASM_MODVERSIONS 1983 bool 1984 default HAVE_ASM_MODVERSIONS && MODVERSIONS 1985 help 1986 This enables module versioning for exported symbols also from 1987 assembly. This can be enabled only when the target architecture 1988 supports it. 1989 1990config MODULE_SRCVERSION_ALL 1991 bool "Source checksum for all modules" 1992 help 1993 Modules which contain a MODULE_VERSION get an extra "srcversion" 1994 field inserted into their modinfo section, which contains a 1995 sum of the source files which made it. This helps maintainers 1996 see exactly which source was used to build a module (since 1997 others sometimes change the module source without updating 1998 the version). With this option, such a "srcversion" field 1999 will be created for all modules. If unsure, say N. 2000 2001config MODULE_SIG 2002 bool "Module signature verification" 2003 select MODULE_SIG_FORMAT 2004 help 2005 Check modules for valid signatures upon load: the signature 2006 is simply appended to the module. For more information see 2007 <file:Documentation/admin-guide/module-signing.rst>. 2008 2009 Note that this option adds the OpenSSL development packages as a 2010 kernel build dependency so that the signing tool can use its crypto 2011 library. 2012 2013 You should enable this option if you wish to use either 2014 CONFIG_SECURITY_LOCKDOWN_LSM or lockdown functionality imposed via 2015 another LSM - otherwise unsigned modules will be loadable regardless 2016 of the lockdown policy. 2017 2018 !!!WARNING!!! If you enable this option, you MUST make sure that the 2019 module DOES NOT get stripped after being signed. This includes the 2020 debuginfo strip done by some packagers (such as rpmbuild) and 2021 inclusion into an initramfs that wants the module size reduced. 2022 2023config MODULE_SIG_FORCE 2024 bool "Require modules to be validly signed" 2025 depends on MODULE_SIG 2026 help 2027 Reject unsigned modules or signed modules for which we don't have a 2028 key. Without this, such modules will simply taint the kernel. 2029 2030config MODULE_SIG_ALL 2031 bool "Automatically sign all modules" 2032 default y 2033 depends on MODULE_SIG || IMA_APPRAISE_MODSIG 2034 help 2035 Sign all modules during make modules_install. Without this option, 2036 modules must be signed manually, using the scripts/sign-file tool. 2037 2038comment "Do not forget to sign required modules with scripts/sign-file" 2039 depends on MODULE_SIG_FORCE && !MODULE_SIG_ALL 2040 2041choice 2042 prompt "Which hash algorithm should modules be signed with?" 2043 depends on MODULE_SIG || IMA_APPRAISE_MODSIG 2044 help 2045 This determines which sort of hashing algorithm will be used during 2046 signature generation. This algorithm _must_ be built into the kernel 2047 directly so that signature verification can take place. It is not 2048 possible to load a signed module containing the algorithm to check 2049 the signature on that module. 2050 2051config MODULE_SIG_SHA1 2052 bool "Sign modules with SHA-1" 2053 select CRYPTO_SHA1 2054 2055config MODULE_SIG_SHA224 2056 bool "Sign modules with SHA-224" 2057 select CRYPTO_SHA256 2058 2059config MODULE_SIG_SHA256 2060 bool "Sign modules with SHA-256" 2061 select CRYPTO_SHA256 2062 2063config MODULE_SIG_SHA384 2064 bool "Sign modules with SHA-384" 2065 select CRYPTO_SHA512 2066 2067config MODULE_SIG_SHA512 2068 bool "Sign modules with SHA-512" 2069 select CRYPTO_SHA512 2070 2071endchoice 2072 2073config MODULE_SIG_HASH 2074 string 2075 depends on MODULE_SIG || IMA_APPRAISE_MODSIG 2076 default "sha1" if MODULE_SIG_SHA1 2077 default "sha224" if MODULE_SIG_SHA224 2078 default "sha256" if MODULE_SIG_SHA256 2079 default "sha384" if MODULE_SIG_SHA384 2080 default "sha512" if MODULE_SIG_SHA512 2081 2082choice 2083 prompt "Module compression mode" 2084 help 2085 This option allows you to choose the algorithm which will be used to 2086 compress modules when 'make modules_install' is run. (or, you can 2087 choose to not compress modules at all.) 2088 2089 External modules will also be compressed in the same way during the 2090 installation. 2091 2092 For modules inside an initrd or initramfs, it's more efficient to 2093 compress the whole initrd or initramfs instead. 2094 2095 This is fully compatible with signed modules. 2096 2097 Please note that the tool used to load modules needs to support the 2098 corresponding algorithm. module-init-tools MAY support gzip, and kmod 2099 MAY support gzip, xz and zstd. 2100 2101 Your build system needs to provide the appropriate compression tool 2102 to compress the modules. 2103 2104 If in doubt, select 'None'. 2105 2106config MODULE_COMPRESS_NONE 2107 bool "None" 2108 help 2109 Do not compress modules. The installed modules are suffixed 2110 with .ko. 2111 2112config MODULE_COMPRESS_GZIP 2113 bool "GZIP" 2114 help 2115 Compress modules with GZIP. The installed modules are suffixed 2116 with .ko.gz. 2117 2118config MODULE_COMPRESS_XZ 2119 bool "XZ" 2120 help 2121 Compress modules with XZ. The installed modules are suffixed 2122 with .ko.xz. 2123 2124config MODULE_COMPRESS_ZSTD 2125 bool "ZSTD" 2126 help 2127 Compress modules with ZSTD. The installed modules are suffixed 2128 with .ko.zst. 2129 2130endchoice 2131 2132config MODULE_DECOMPRESS 2133 bool "Support in-kernel module decompression" 2134 depends on MODULE_COMPRESS_GZIP || MODULE_COMPRESS_XZ 2135 select ZLIB_INFLATE if MODULE_COMPRESS_GZIP 2136 select XZ_DEC if MODULE_COMPRESS_XZ 2137 help 2138 2139 Support for decompressing kernel modules by the kernel itself 2140 instead of relying on userspace to perform this task. Useful when 2141 load pinning security policy is enabled. 2142 2143 If unsure, say N. 2144 2145config MODULE_ALLOW_MISSING_NAMESPACE_IMPORTS 2146 bool "Allow loading of modules with missing namespace imports" 2147 help 2148 Symbols exported with EXPORT_SYMBOL_NS*() are considered exported in 2149 a namespace. A module that makes use of a symbol exported with such a 2150 namespace is required to import the namespace via MODULE_IMPORT_NS(). 2151 There is no technical reason to enforce correct namespace imports, 2152 but it creates consistency between symbols defining namespaces and 2153 users importing namespaces they make use of. This option relaxes this 2154 requirement and lifts the enforcement when loading a module. 2155 2156 If unsure, say N. 2157 2158config MODPROBE_PATH 2159 string "Path to modprobe binary" 2160 default "/sbin/modprobe" 2161 help 2162 When kernel code requests a module, it does so by calling 2163 the "modprobe" userspace utility. This option allows you to 2164 set the path where that binary is found. This can be changed 2165 at runtime via the sysctl file 2166 /proc/sys/kernel/modprobe. Setting this to the empty string 2167 removes the kernel's ability to request modules (but 2168 userspace can still load modules explicitly). 2169 2170config TRIM_UNUSED_KSYMS 2171 bool "Trim unused exported kernel symbols" if EXPERT 2172 depends on !COMPILE_TEST 2173 help 2174 The kernel and some modules make many symbols available for 2175 other modules to use via EXPORT_SYMBOL() and variants. Depending 2176 on the set of modules being selected in your kernel configuration, 2177 many of those exported symbols might never be used. 2178 2179 This option allows for unused exported symbols to be dropped from 2180 the build. In turn, this provides the compiler more opportunities 2181 (especially when using LTO) for optimizing the code and reducing 2182 binary size. This might have some security advantages as well. 2183 2184 If unsure, or if you need to build out-of-tree modules, say N. 2185 2186config UNUSED_KSYMS_WHITELIST 2187 string "Whitelist of symbols to keep in ksymtab" 2188 depends on TRIM_UNUSED_KSYMS 2189 help 2190 By default, all unused exported symbols will be un-exported from the 2191 build when TRIM_UNUSED_KSYMS is selected. 2192 2193 UNUSED_KSYMS_WHITELIST allows to whitelist symbols that must be kept 2194 exported at all times, even in absence of in-tree users. The value to 2195 set here is the path to a text file containing the list of symbols, 2196 one per line. The path can be absolute, or relative to the kernel 2197 source tree. 2198 2199endif # MODULES 2200 2201config MODULES_TREE_LOOKUP 2202 def_bool y 2203 depends on PERF_EVENTS || TRACING || CFI_CLANG 2204 2205config INIT_ALL_POSSIBLE 2206 bool 2207 help 2208 Back when each arch used to define their own cpu_online_mask and 2209 cpu_possible_mask, some of them chose to initialize cpu_possible_mask 2210 with all 1s, and others with all 0s. When they were centralised, 2211 it was better to provide this option than to break all the archs 2212 and have several arch maintainers pursuing me down dark alleys. 2213 2214source "block/Kconfig" 2215 2216config PREEMPT_NOTIFIERS 2217 bool 2218 2219config PADATA 2220 depends on SMP 2221 bool 2222 2223config ASN1 2224 tristate 2225 help 2226 Build a simple ASN.1 grammar compiler that produces a bytecode output 2227 that can be interpreted by the ASN.1 stream decoder and used to 2228 inform it as to what tags are to be expected in a stream and what 2229 functions to call on what tags. 2230 2231source "kernel/Kconfig.locks" 2232 2233config ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE 2234 bool 2235 2236config ARCH_HAS_SYNC_CORE_BEFORE_USERMODE 2237 bool 2238 2239# It may be useful for an architecture to override the definitions of the 2240# SYSCALL_DEFINE() and __SYSCALL_DEFINEx() macros in <linux/syscalls.h> 2241# and the COMPAT_ variants in <linux/compat.h>, in particular to use a 2242# different calling convention for syscalls. They can also override the 2243# macros for not-implemented syscalls in kernel/sys_ni.c and 2244# kernel/time/posix-stubs.c. All these overrides need to be available in 2245# <asm/syscall_wrapper.h>. 2246config ARCH_HAS_SYSCALL_WRAPPER 2247 def_bool n 2248